The Section on Neurophysiology employs the multidisciplinary methods of behavioral neurophysiology in nonhuman primates. Our studies are aimed at illuminating two major problems in neurobiology: (a) the specialized functions of the different frontal neocortical areas and (b) the neural network underlying one particular kind of flexible behavior, the arbitrary mapping of visual antecedents onto motor consequents. Our studies have focused on the neuronal network underlying one form of visuomotor learning, which we have termed arbitrary visuomotor mapping. In the past year, we have provided and reviewed evidence for a role of hippocampus and prefrontal cortex in such mappings (Wise and Murray, 1999; Murray, Bussey and Wise, 2000). We have published the first systematic report of changes in activity in motor areas of cortex during traditional visuomotor mapping (as contrasted with arbitrary visuomotor mapping) (Wise, Moody, Blomstrom and Mitz, 1998) and have provided the first evidence for a signal in prefrontal cortex reflecting a behavior-guiding rule (White and Wise, 1999). We have also explored the functional significance of the various patterns of activity observed in frontal cortex during instructed delay and matching-to-sample tasks (Moody and Wise, 2000a, b). Lebedev and Wise (2000) showed the presence of oscillatory features in motor cortical activity. Our proposal for future research involves five projects, each aimed at testing a specific and central hypothesis about either frontal lobe function or the neural network underlying arbitrary visomotor mapping. We propose: 1. To test the hypotheses that striatal cells change activity later in arbitrary visuomotor learning (relative to premotor and prefrontal cortex) and that striatal activity reflects the context for actions, as predicted in the model of Wise, Gerfen and Murray (1996). 2. To test the hypothesis that prefrontal cortical areas abstract information to compute novel combinations of information. 3. To test the hypothesis that activity in the dorsolateral prefrontal cortex reflects attentional information properties in addition to or rather than the mnemonic functions commonly assumed. 4. To test the hypothesis that medial frontal areas reflect the direction of action based on internal computations (premised on recent history), whereas lateral frontal areas reflect the direction of action based on external cues (which vary from trial to trial). 5. To test the hypothesis that the hippocampal system computes generalized arbitrary mappings rather than those restricted to spatial or ideothetic information. - motor systems premotor prefrontal conditional motor learning cortex